Perfect test accuracy does not imply symbolic-level logical reasoning: GPT-5 gives 100% right syllogism decisions with 25 wrong explanations, and the image-based SupEN stalls at 50-97.8% per mood.
arXiv:2410.14399 (2025) [cs.CL]
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Data-driven Machine Learning Cannot Reach Symbolic-level Logical Reasoning -- The Limit of the Scaling Law
Perfect test accuracy does not imply symbolic-level logical reasoning: GPT-5 gives 100% right syllogism decisions with 25 wrong explanations, and the image-based SupEN stalls at 50-97.8% per mood.